/Glossary
AWS, DevOps and generative AI glossary
Straight definitions, when each option makes sense and the most common mistakes. Written by the team that designs and runs AWS platforms every day.
AWS
- Amazon EKSAmazon EKS (Elastic Kubernetes Service) is the AWS managed service for running Kubernetes without operating the control plane. AWS handles control plane availability and upgrades, and you deploy your workloads with standard Kubernetes tooling.
- Amazon ECSAmazon ECS (Elastic Container Service) is the fully managed AWS container orchestrator. It runs and scales containers on AWS Fargate, with no servers to manage, or on EC2 instances, with no control plane to maintain.
- AWS LambdaAWS Lambda is the AWS serverless compute service that runs code in response to events without you provisioning or managing servers. It scales automatically and charges per request and for execution time based on allocated memory.
AI
- Amazon BedrockAmazon Bedrock is a fully managed AWS service that gives you access to foundation models from providers like Anthropic, Meta, Mistral AI and Amazon through a single API. It includes the tooling to build generative AI applications and agents with enterprise-grade security and privacy.
- RAGRAG (Retrieval-Augmented Generation) is a technique that improves a language model's answers by retrieving relevant information from your own sources and passing it into the prompt as context. The model can then answer with current, citable data without being retrained.
- AI agentsAI agents are systems that use a language model to plan and carry out multi-step tasks, calling tools, APIs and data sources on their own. Unlike a chatbot, they take actions to reach a goal.
- LLMOpsLLMOps is the set of practices for building, deploying, evaluating and operating applications based on language models in production. It covers prompt versioning, quality evaluation, observability, cost control and security.
- MLOpsMLOps is the set of practices that applies DevOps principles to the machine learning lifecycle, from data preparation and training to deploying and monitoring models in production. The goal is models that are reproducible, auditable and easy to update.
- ClaudeClaude is the family of large language models developed by Anthropic. It can reason over text, code, and images, and it can use tools. On AWS, you access it through Amazon Bedrock, inside your own account and under your IAM controls.
- Claude CodeClaude Code is Anthropic's agentic coding tool that reads your repository, edits files, runs commands, and proposes changes from natural-language instructions. It works in the terminal, in your IDE, and in CI pipelines.
- Model Context Protocol (MCP)Model Context Protocol (MCP) is an open standard, introduced by Anthropic, that defines how AI applications connect to external tools, data, and services. It lets you write an integration once and use it from any compatible client.
DevOps
- DevOpsDevOps is a way of working that brings development and operations together to ship software frequently, safely and repeatably. It combines a shared culture, an automated delivery lifecycle and continuous measurement of results.
- Platform engineeringPlatform engineering is the discipline of designing and running a self-service internal platform so development teams can build, deploy and operate software without filing tickets. The platform is treated as a product with internal users.
- Infrastructure as codeInfrastructure as code (IaC) is the practice of defining servers, networks, permissions and cloud services in version-controlled files instead of configuring them by hand. That way infrastructure is reviewed, tested and deployed just like software.
- TerraformTerraform is an infrastructure as code tool from HashiCorp that defines cloud resources in declarative HCL files and creates, updates or deletes them based on a plan. It works with AWS and hundreds of other providers.
- CI/CDCI/CD is the set of continuous integration and continuous delivery or deployment practices that automates building, testing and releasing every code change. It lets you ship small changes to production frequently and reliably.
- GitOpsGitOps is an operating model in which Git is the single source of truth for the desired state of infrastructure and applications. An agent continuously compares that state with the live environment and reconciles it automatically.
- SRESRE (Site Reliability Engineering) is a discipline that applies software engineering practices to operating systems so they are reliable and scalable. It sets measurable reliability targets and reduces manual operational work through automation.
- ObservabilityObservability is the ability to understand a system's internal state from the data it emits, mainly metrics, logs and traces. It lets you answer new questions about a problem without having to deploy additional code.
- AgentOpsAgentOps is the set of practices for taking AI agents to production and running them safely: which tools and permissions each agent has, how its decisions are logged, how its quality is evaluated, what each run costs, and where a person signs off.
Architecture
- ServerlessServerless is a cloud computing model in which the provider manages servers, scaling and availability, and you pay only for actual usage. It lets you focus on code and business logic instead of infrastructure.
- Landing zoneA landing zone is a secure, well-governed multi-account AWS environment that serves as the foundation for your workloads. It defines account structure, identity, networking, centralized logging, and security controls from the start.
- AWS Well-ArchitectedAWS Well-Architected is the AWS best-practice framework for designing and running secure, reliable, efficient, and cost-effective workloads. It is organized into six pillars and provides concrete questions to assess an architecture and surface risks.
- Multi-tenant SaaSMulti-tenant SaaS is a software-as-a-service application in which multiple customers, called tenants, share the same platform and codebase while their data and permissions stay isolated from each other. It lets you scale to many customers at a lower operating cost.
Security
- AWS Control TowerAWS Control Tower is an AWS service that sets up and governs a multi-account landing zone based on best practices. It automates account creation, applies security and compliance controls, and centralizes visibility across your whole organization.
- Zero TrustZero Trust is a security model that grants no implicit trust to any user, device, or network, even inside the perimeter. Every access request is explicitly verified based on identity, context, and least privilege.
Cost
- FinOpsFinOps is a cloud financial management practice that brings engineering, finance and the business together to make data-driven spending decisions. The goal is for every dollar spent on the cloud to deliver measurable value.
- Savings PlansSavings Plans are an AWS pricing model that offers lower rates in exchange for committing to a consistent hourly spend for one or three years. The discount applies automatically to eligible usage, without tying you to a specific instance.
Next step
What should we build next on AWS?
30 minutes with an engineer. Concrete next steps, whether or not you work with us.